explain
/explain
Explains code or a process as a drawing that builds itself, in plain words for people who don't read code. Use when someone asks "what happens when I click X", "how does this part of my app work", "walk me through this flow", or asks to explain, draw or visualise a process, pipeline, agent loop or business flow.
- Author
- nicodiansk ↗
- Repository
- draw-it-out ↗
- Licence
- MIT
- Last repository change
- 13.11.2026
Assessment and security
- Files in this package
- 15
- Scripts
- 7
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 13.11.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
- 7 scripts in this package. Review them before installing, as Claude may execute them.
Scripts: engine/capture.py, engine/render.py, examples/shop/api/pay.js, examples/shop/app/pay-button.js, examples/shop/lib/card.js, examples/shop/lib/db.js, examples/shop/tests/db.test.js
What it does and when to use it
Explains code or a process as a drawing that builds itself, in plain words for people who don't read code. Use when someone asks "what happens when I click X", "how does this part of my app work", "walk me through this flow", or asks to explain, draw or visualise a process, pipeline, agent loop or business flow.
Triggers from the description
Original author description
Explains code or a process as a drawing that builds itself, in plain words for people who don't read code. Use when someone asks "what happens when I click X", "how does this part of my app work", "walk me through this flow", or asks to explain, draw or visualise a process, pipeline, agent loop or business flow.
How it works
- Pick the mode
- Rules for every sheet
- Render
- Show it
SKILL.md preview
---
name: explain
description: Explains code or a process as a drawing that builds itself, in plain words for people who don't read code. Use when someone asks "what happens when I click X", "how does this part of my app work", "walk me through this flow", or asks to explain, draw or visualise a process, pipeline, agent loop or business flow.
---
# Draw it out
You explain one thing as a **sheet**: a drawing that builds itself piece by piece. The reader
built software with AI and can't read the code, so every word on the sheet is plain.
## Pick the mode
- **walk**: the question is what *their own code* does when they do something ("what happens
when I click Pay?", "where does my signup form go?"). Read [walk.md](walk.md).
- **sheet**: the question is how a process works (a pipeline, an agent loop, a business flow,
a document they shared). Code is optional. Read [sheet.md](sheet.md).
Not sure which? Ask one question: "Should I follow it through your code, or draw the process?"
Read only the mode's file, then [schema.md](schema.md) for the JSON you write.
## Rules for every sheet
- Write in the asker's language, whatever it is: `text`, `label`, `title` and `lang`. For a
language other than English or Italian, also fill `ui` (schema.md).
- Plain words, short sentences. Explain each technical term the first time it appears
("the server: the computer that runs your app's code").
- Never describe what you didn't see. A gap goes in `unsure`, said plainly.
- Code reaches the sheet only through `render.py`: you write `{"file", "lines"}`, never the code.
- Read-only: never edit the project, submit a form, sign in or pay.
## Render
Write the JSON (and any screenshots) to a folder outside the project, such as your scratchpad
or a temp folder, then run:
```
python3 <this skill's folder>/engine/render.py <folder>/walk.json --root <project folder> --out <folder>/walk.html
```
`--root` is the folder that code paths start from; screenshot paths start from the JSON's folder.
Exit 1 prints one line per problem: `field → problem → how to fix`. Fix the JSON and rerun,
at most three times. Still refused: stop and tell the user, in plain words, what is missing.
## Show it
- An Artifact tool is available: publish the HTML file as an artifact.
- Claude Code without one: give the path and open it (`open` on macOS, `xdg-open` on Linux,
`start` on Windows).
- claude.ai without one: share the HTML file.
Then two or three lines: what the sheet shows, the one piece to look at, and any gap from `unsure`.
Files in this package (15)
- SKILL.md 2 KB
- schema.md 3 KB
- sheet.md 1 KB
- walk.md 2 KB
- engine/capture.py 4 KB
- engine/render.py 18 KB
- engine/template.html 26 KB
- examples/sheet.json 1 KB
- examples/walk.json 2 KB
- examples/shop/api/pay.js 1 KB
- examples/shop/app/pay-button.js 1 KB
- examples/shop/lib/card.js 1 KB
- examples/shop/lib/db.js 1 KB
- examples/shop/shots/checkout.svg 1 KB
- examples/shop/tests/db.test.js 1 KB
Versions
| Repository | Score | Licence |
|---|---|---|
| This page · nicodiansk/draw-it-out | 100 | MIT |
| crbnos/carbon | 100 | AGPL-3.0 |
| PerilloAI/flowtrace-plugin | 100 | Apache-2.0 |
| mickeyyaya/evolve-loop | 100 | Apache-2.0 |
| zenobi-us/dotfiles | 100 | MIT |
| imgly/agent-skills | 100 | MIT |
| marcel-bich/marcel-bich-claude-marketplace | 100 | MIT |
| ramboz/jig | 100 | MIT |
| ramboz/jig | 100 | MIT |
Database snapshot: 13.11.2026 · package version 6dccc3e5e073 (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 90explain-code
Explain how a piece of code works (a function, file, module, or a flow like "what happens when a user signs up") with file and line references, traced from the actual code. Use when asked "explain this", "how does X work", "what does this function do", "walk me through", or "where is X handled". - 90ownership
Explain code and decisions so the user fully owns them - plain words, analogies, file:line walkthroughs, rejected alternatives, real references, and a way to verify it themselves. Use on every task that writes, changes, reviews or explains code, and when the user says… - 100explain-flow
Use when the owner wants a flow of this codebase explained as a story ("how does X run", "walk me through the flow of Y"), with every step checked against the code. - 100code-slowmo
Explains code with an interactive visual page instead of a wall of text. It builds one HTML file with a clickable map of the system, a step-by-step timeline of what each part tells the others, and a line-by-line run of the real code that shows changing values, a log and a mock… - 90ctx-explain
Explain code for someone new to the project. Use when the user asks 'what does this do', 'explain this', or wants to understand unfamiliar code. - 85atlas-explain
Explains a piece of the work to a person who was not there: a mechanism in plain words for someone who does not code, a decision with its options and its cost, a set of measurements with what they do and do not prove. Use for "atlas explain", "explain this simply", "why did we…